A Standardised Procedure
for Surveillance and Monitoring
European Habitats
R.G.H. Bunce1, J. Brandt2, G. de Blust3, R. Elena-Rossello4,
G.B. Groom5, L. Halad6, G. Hofer7, D.C. Howard8, R.H.G. Jongman1,
P. Kovar9, M.J. Metzger10, C.A. Mucher1,
E. Padoa-Schioppa11, D. Paelinckx3, A. Palo12, M. Perez-Soba1,
I. Ramos13, P. Roche14, H. Skanes15 and T. Wrbka16
1Alterra
Green World Research, Wageningen, The Netherlands; ; 2Roskilde University,
Department of Environmental, Social and Spatial Changes, Roskilde, Denmark; 3Institute of
Nature Conservation and Forestry Management, Brussels, Belgium; 3Polytecnic University of
Madrid, Department of Forestry, Madrid, Spain; 5NERI, Department of Wildlife Ecology and
Biodiveristy, Roende, Denmark; 6Institue of Landscape Ecology, Slovak Academy of
Sciences, Bratislava, Slovakia; 7FAL, Zurich, Switzerland; 8CEH Lancaster, UK; 9Charles
University, Faculty of Science, Prague, Czech Republic; 10Wageningen University,
Department of Environmental Systems Analysis, Wageningfen, The Netherlands; 11University
of Milan, Department of Environmental Science, Milan, Italy; 12University of Tartu, Tartu,
Estonia; 13CESUR, Lisbon, Portugal; 14University of Aix en Provence, Marseille, France;
15Stockholm
University, Department of Physical Geography, Stockholm, Sweden; 16University
of Vienna, Institute of Conservation Biology, Vegetation Ecology and Landscape Ecology,
Vienna, Austria;
Corresponding author: bob.bunce@wur.nl
Contents
Abstract ...................................................................................................................................... 3
Introduction ................................................................................................................................ 4
The Conceptual Principles ......................................................................................................... 6
The Surveillance System ........................................................................................................... 8
Validation ................................................................................................................................. 14
Discussion ............................................................................................................................... 18
References .............................................................................................................................. 20
Abstract
There are well defined scientific and policy requirements for a practical, transmissible, and
reproducible procedure for surveillance and monitoring of European habitats. A procedure is
described that will satisfy these requirements and which can provide the necessary data that
are currently lacking. Rigorous rules are required; otherwise changes from baseline records
cannot reliably be separated from background noise. The procedure is based on classical
plant Life Forms, used in biogeography since the nineteenth century, and on the underlying
statistical correlation between these Life Forms and the environment. This relationship has
been validated statistically and the procedure has also been tested in the field in all European
environmental zones. 130 General Habitat categories are described and these are enhanced in
the field by recording environmental, site and management qualities to enable flexible
database interrogation. The same categories are applied using rules, with appropriate
qualifiers, to areal, linear and point features, so that for the first time integrated reporting of
variation of habitats at the landscape level can be carried out. It therefore incorporates
landscape ecological principles such as landscape diversity and will enable both connectivity
and fragmentation to be assessed at the detailed landscape level.
Keywords: field recording, surveillance, habitat, monitoring, stratification, biodiversity, Life
Forms
Introduction
The current primary policy requirement in NATURA 2000 sites is to assess the favourable
conservation status and extent of selected species and habitats. This requires a reliable,
comparable and repeatable method for monitoring changes in species and habitats, both
within and outside, NATURA 2000 sites. However, monitoring is also needed for other
initatives, such as indicators of sustainable development and biodiversity action plans.
Currently not only are consistent data not available to answer these requirements, but also a
standardised rule based procedure is not available. The approach described in the present
paper fills this gap and is based on the BioHab project carried out in the EU Fifth Framework
programme (EVK2-CT-2002-20018). Whilst the core of the procedure concerns rules and
instructions for consistent field recording, it is essential that they are linked to a spatial
framework for the whole of Europe. [Ramon’s diagram here of pstial and temporal
issues]. Such a framework is therefore integral to the methodology as it provides a means to
extend the detailed samples needed to assess habitats in the field to European estimates. A
summary of the framework is also therefore provided. The temporal dimension is added by
describing the monitoring procedure.
Field recording has been at the core of ecology since its inception as a recognisable science.
Long-term monitoring is more common in species oriented approaches e.g. birds and carabid
beetles (Den Boer and Van Dijk 1994). The development of vegetation science as part of
ecology has been mainly descriptive and based on the selection of homogenous stands of
vegetation, usually relatively undisturbed (Braun-Blanquet 1932; Tuxen 1937). Such work is
not designed for long-term monitoring, although the individual records are suitable, if they are
relocateable (e.g. Grabherr et al. 1994).
Long-term vegetation monitoring studies are
restricted. For example, in the UK (Bunce et al. 1993) found under five case studies of longterm vegetation monitoring. Similarly, quantitative ecologists, e.g. Greig-Smith (1964) were
largely concerned with technical, as opposed to actual practical problems. Bunce and Shaw
(1973) described a standardised procedure, which was applied to British woodlands in 1971
and subsequently repeated in 2003 (Kirby et al. 2005) demonstrating that statistical rigour is
essential for long-term monitoring. However, all these approaches do not involve recording
at the landscape level i.e. involving complexes of habitats, as described by Sheail and Bunce
(2003) and Fjellstad et al. (2001). Any procedure must recognise and utilise this complexity
– hence the development of the present approach. In contrast, most European approaches to
the assessment of habitats either ignore these requirements, or deliberately avoid landscape
complexitiy by selection of sites that are considered homogenous. In the 1950’s and 60’s, the
development of the ecosystem concept was restricted, albeit not explicitly, to concepts of
vegetation classification. However, in the 1980’s it gradually became recognised that, whilst
habitats had strong links to vegetation classes, they could also be independent. This was
partly because animal ecologists found that vegetation structure often overrode vegetation
classes but also partly because some widely recognised habitats were not directly linked to
traditional vegetation associations. This situation has recently been recognised by Rodwell et
al. (2003) showing that the match between the European Habitat Classification (EUNIS)
(Davis and Moss 2002) and vegetation assemblages is often indistinct.
In the 1980’s therefore, habitat mapping progressively became a separate exercise from
recording vegetation alone because strategic conservation priorities did not necessarily
involve the distinction between vegetation associations. For example, the small biotope
project in Denmark (Agger and Brandt (1988) monitored changes in small landscape patches
in intensively farmed landscapes, with minimal relationships with vegetation associations.
Similarly, the Phase 1 Habitat Survey in England (JNCC (1990)) enabled rapid mapping to be
carried out over large areas to provide a strategic basis for determining conservation priorities
albeit at a low level of detail and consistency. Similarly, an examination of the development
of the Countryside Survey in the UK (Haines-Young et al. (2000)) shows that although it
initially concentrated on vegetation in 1978, by 2000 the reporting of status and change was
based on 19 Broad Habitats. Whilst this project has shown the essential role of vegetation
records in determining quality (i.e. favourable conservation status) it has also demonstrated
that habitats are convenient for reporting. Furthermore, the habitat names are more often than
vegetation classes understandable by policy makers. . Landscapes usually contain complexes
of habitats whereas at the habitat level below contain mixtures of vegetation associations.
The list of CORINE Biotopes (Devillers et al. 1991) was derived from expert group
discussions. It was largely based on vegetation classes and mainly concerned semi-natural
habitats. Both Annex 1 of the European Habitats and Species Directive and the subsequent
Palaearctic classification (De Villers and Devillers-Terschuren (1996) have strong links to the
CORINE Biotope Classes. More recently, EUNIS (Davies and Moss (2002) whilst still
maintaining strong links to the previous classifications, also introduced classes for artificial
and highly disturbed situations. The present project originally intended to develop rules for
the EUNIS classes but concluded that many terms used in the key e.g. montane and subMediterranean were not sufficiently well defined for determining the classes in the field and
for subsequent monitoring.
Accordingly, the approach developed adopted traditional
scientific principles in developing General Habitat Categories based on plant Life Forms,
appropriate for monitoring and reporting consistently as the European scale. The present
paper first describes these principles and the validation process accompanying them. The
surveillance system is then described with a summary of the principal rules and the method of
recording qualifiers to convey information on drivers and descriptive characters. Finally, the
environmental framework for relating the necessary detailed samples to the whole population
is described. Additional details on structure are also provided to provide better links of in situ
to remote sensed information.
The Conceptual Principles
Originally plant Life Forms were identified as valuable in the project because they provided
rules to separate grassland, scrub and forest categories using rules that could be applied
consistently in the field. However, during the project it became clear that Life Forms
provided a means of transcending species and enabling consistent recording of habitats to be
undertaken. It is generally recognised that at the continental level biomes need to be defined
in terms of the physiognomy of the dominant species and that Life Forms are necessary
because individual species are too limited to encompass widely dispersed geographical
locations (Woodward & Rochefort 1991). It was also realised in the project that the adoption
of Life Forms would provide links between European categories and other studies of global
change that use biomes based largely on Life Forms e.g. Mediterranean scrub in the western
USA, Chile, South Africa and Australia. By contrast, many criteria have been used to
construct existing European habitat classifications e.g. species, geographical distribution,
vegetation classes and environmental factors. It was therefore decided to use Life Forms as
the sole criterion for determining the primary General Habitat Categories (GHC’s) so that
they can be recorded directly in the field but can also be sufficiently robust to be used to link
existing datasets which have been collected for monitoring. The basis of the GHC’s is the
classification of plant Life Forms produced by Raunkiaer (1907; 1934). The underlying
scientific hypothesis is that habitat structure is related to the environment on a European
Scale, or even locally if there is a sufficiently wide range of conditions.
The application of the Environmental Stratification (Metzger et ai. 2005; Jongman et al.
2006) provides a sampling framework linked to climate, topography and geographical
location, which can also be tested statistically. Various floras were consulted, especially
Clapham et al. (1952) and Pignatti (1982) to determine at what level to treat Life Forms as
some recent floras e.g. Oberdofter et al. (1990) give highly detailed categories. However, as
Raunkiaer (1934) originally emphasized, the more detailed breakdown of Life Forms, loses
the strong relationship with climate.
Eventually, it was decided to use 16 Life Forms
(Herbaceous and Tree/scrub, see Table 1) with the plant height ranges taken from more recent
literature Ie.g. Castri et al. (1981); Quetzal & Barber (1982).
The main problem was
however, with Gramineae, Cyperaceae and Juncaceae, where many species have rhizomes,
which are primarily for vegetational reproduction not for perennation.
There are also
differences between the attribution floras of Life Forms as well as difficulties in the
determination of the actual position of the rhizomes or stolons in the field. It was therefore
decided to group these three taxa together as “caespitose hermicryptophytes”. Further details
and examples of the species in the 16 Life Forms are given in Bunce et al. (2005). It is also
recognised that some species are sufficiently plastic to adapt to several habitats, e.g.
Ranunculus aquatilis; in which case the environmental conditions present at the site, as
described below should be used to determine whether, for example, it is in aquatic or
waterlogged conditions.
Another aspect of plasticity relates to woody species as shown in Table 2 which respond to a
range of environmental and management pressures. These species can occur in lower than
optimum height categories because:

They have been heavily grazed

They have been burnt

They are regenerating

They are in highly exposed or extreme environments
The first three categories are transitional and shifts can take place according to changes in
external pressures e.g. fire or felling. The fourth is a climax state. The only way to provide
consistent data is to record the actual hights of the tree and shrub cover in the field, because
otherwise the potential height is a matter of judgement. There is a functional relationship
between the Life Forms and pathways of change over time e.g. from tall shrubs 2-5m to forest
trees over 5m which are then quantifiable, as in the flow diagrams beween habitats given by
Haines-Young et al. (2000).
Land associated with built structures and routes of communication (termed urban in a broad
sense) and crops cannot be defined solely in terms of Life Forms as they are primarily land
uses. However, for policy and practial reasons it is essential that such land is separated from
other land covers that are mainly in agricultural or forest use. Hence, these two categories
have been separated at the first level of the hierarchy (Bunce et al. 2005), together with bare
land as shown in Table 1. However, within both the former categories, subsequent divisions
are then based on life forms at the second level of Table 1.
These are termed super
categories. A major problem in habitat classifications is how to determine the number of
classes. In some habitat classifications by e.g. Fernandez (2003) there are almost 1000
classes and in EUNIS there are 350 at level three. It was therefore decided that below the first
tier of five super categories all possible combinations of Life Forms should be included, even
although some will be rare (Table 1). This procedure has provided a statistical rule for
determining the number of General Habitats Categories (HGC’s) and results in 130 covering
the pan-Euroepan region, except Turkey. Other Life Forms e.g. tall succulents would have to
be included for a larger region but at present they are included as qualifiers.
Other
qualifications to be GHC’s are discussed below. The principal reason behind the GHC’s is
that they enable the primary decision on the habitat categories to be made in the field and the
rules and instructions also make them appropriate for monitoring. A worked example of
monitoring national change using a similar level of categories of habitats is given by HainesYoung et al. (2000).
The Surveillance System
General Habitat Categories
The General Habitat Categories are designed for recording habitats and providing links to
their classifications. The General Habitat Categories are based mainly on Life Forms with
added detailed information on environment, site, management and species composition. They
are designed for consistent recording as described by Bunce & Shaw (1973) and are based on
the same principle of using explicit rules. The success of these rules in long-term monitoring
has been shown by Kirby et al. (2005).
Because GHC’s are fundamental ecological
categories, the transfer between them can be readily defined as shown in Figure 3 and as
explained above. The working definition of “habitat” developed by the BioHab project
(Bunce et al. 2005) is as follows: “An element of land that can be consistently defined
spatially in the field in order to define the principal environments in which organisms live”.
General Habitat Categories (GHC’s) have been developed in order to enable consistent
recording of habitats by defining basic mapping units that can then be re-surveyed after a
given time interval (Table 3). It is preferable to carry out preparatory work on delineation of
the major elements within the survey area from the aerial photographs and related material,
e.g. cadastral maps, to assist the mapping process, otherwise it has been shown that
differences in base maps between countries do not enable consistent mapping. Details of the
practical mapping procedure are given in Bunce et al. (2005).
The recording procedures are determined in advance to ensure standarised data as shown in
Table 4. All major decisions should be made in the field rather than being postponed to the
laboratory to improve consistency. Data mining through database management methods can
however be used to extract other data, e.g. calculation of slope angles, aspect and height of
cliffs, land units of ownership.
For survey, the recording of the GHC’s should be made in a time window either side of the
period of maximum biomass, i.e. as close as possible to the height of the growing season.
The monitoring should be carried out as close as possible at the same date (Barr et al. 1993)
as it has been shown that date is the main source of error. This window is likely to be fore
maximum biomass in the Mediterranean, but after in Scandinavia. The extent of the window
needs to be set by region, using local phonological information and differs between
Environmental Zones, Environmental Strata and countries. Some local flexibility may be
required for annual variations in seasons and weather.
Environmental terms must be considered at a contintental scale e.g. “dry” in Scotland may be
“mesic” compared with southern Italy. This means that no continent wide survey can be
carried out without adequate field training for all surveyors to ensure that terms are fully
understood. Combined teams of two people, probably with a botanist and an experienced
mapper, are needed to ensure that optimal expertise is available. Quality assurance and
control are essential and should be carried out regularly with standard protocols, as described
by Barr et al. (1993) and Kirby et al. (2005).
In the BioHab project, it was decided to apply the same recording format to be used for areal,
linear and point elements. This procedure not only uses fewer categories to be used but also
enables reporting to be more readily understood. It also helps express the contrasts between
different types of landscapes where areal and linear features make contrasting contributions to
diversity. Project qualifiers are recorded in seven sequential fields, as described in Table 5.
These can consist of data recorded in the field as well as through database management. In
total the recording form has an alpha identifier and eight subsequent recording fields (Table
5).
The determination of the GHC is based upon a sequence of four dichotomous divisions
related to a set of five super categories as shown above, which determine the set of Life
Forms that can be used to identify the appropriate GHC. The first decision concerns whether
the element is Urban, the second whether it is a Crop, the third whether it is Sparsely
Vegetated and the fourth whether it is Trees or Shrubs (Table 6). Rules have then been added
for further divisions in all super categories and habitat categories including percentage
criteria. New areal or linear elements are separated from adjacent or surrounding areal
elements based on pre-defined rules as described by Bunce et al. (2005).
Environmental Qualifiers
Environmental and global qualifier codes are essential information for detailing GHC’s; they
are therefore entered into the second field of the habitat recording sheets for areal and for
linear elements and they express variation between elements that have the same GHC. They
are not applied to urban/constructed, crop or sparsely vegetated elements.
The moisture categories are based on Pyatt (1999) and are defined in a European gradient as
shown in Table 7.
The second environmental gradient is determined on nutrient levels in the soil. These were
originally developed for Central Europe by Ellenberg (1992) using expert knowledge of the
environmental amplitude of individual species. The species have been recalibrated for Britain
(Hill
et
al.
####)
and
are
available
on
the
web
(htt://science.ceh.ac.uk/products_services/software/mavis.htm). However, Ellenberg values
are not available for some regions, so local experience of the ecological amplitude of species
may be needed, especially in the Mediterranean.
Ellenberg values are available for fertility, acid/basic gradient and salinity. Fertility is often
localised along landscape elements e.g. rivers and around feeding troughs. For ease of
recording a matrix as shown in Table 8 is used to determine codes. Indicator species can be
used to identify such elements e.g. Urtica spp., Stellaria media, Galium aparine, Stachys
suylvatica and Rumex alpinum. The two highest levels of Ellenberg fertility values are
combined because lower levels are too difficult to record in the field without a full species
list.
Indicator species can also be used for the acic/basic gradient values and can be supported by
knowledge of the soil/rock type, landscape context and if necessary, by testing the soil
reaction.
Salinity can be assessed by the presence of halophytes e.g. Salicornia spp.,
Puccinellia spp and Spartina spp. Care is needed with some species e.g. Armeria maritime
and Plantago maritime as they also grow in mountains away from saline conditions. Brackish
conditions can be determined from the landscape context and the presence of some species
that are some degree tolerant of salt e.g. Agrophyron repens and Zannichelia palustris.
Whilst there is an element of judgement in assessing the position of a given element on these
gradients, they are mainly stable and when monitoring changes should only be recorded if
there is supporting evidence e.g. flooding for a different position in the matrix. For all cells in
the matrix the overall balance of species should be used, not individual indicators. Other
codes are also provided, as described by Bunce et al (2005).
Environmental qualifiers are determined for each mapped element except for the urban,
sparsely vegetation and crop super categories.
Site qualifiers are recorded in field three including factors such as geomorphologic features
and coastal attribute. Management qualifiers are grouped in convenient sections e.g. forestry
and reacreation and are designed to given information as potential causes of change.
The detailed information of the areal and linear elements, the main plant and crop species
associated with each recorded alpha code are recorded. Life Forms with a cover of at least
10% are also recorded. The species that constitute at least 30% cover of the vegetation (as
seen in vertical perspective) of each Life Form that has been recorded in the first column of
field five are recorded, as well as the percentage cover. Three further fields are provided for
pan-European habitat classifications e.g. EUNIS, local habitat classifications e.g. MorilloFernandez (2003) and for phytosociological associations. All the above data can be recorded
on a hand-held computer and a pilot system was developed in the BioHab project.
Mapping Units and Areal Elements
One of the main problems in determining the number of samples is that a given habitat often
occurs at different scales in contrasting landscapes. The recording procedure therefore needs
to reflect such heterogeneity. In the present paper, it is recognised that the optimum size of
the sampling unit depends on the objective of study (Lambert 1972). The choice of scale is
therefore a compromise between sampling many small units, as opposed to fewer larger units.
As discussed by Bunce et al. (1996) it costs more to sample many small units although they
may give more precise estimates (Gallego 2002). More systematic inclusion of variations due
to management might have been an argument for larger units (Brandt et al. 2002), but the 1
km square is a convenient compromise sampling unit for many purposes as there is no
optimal size for all the habitats and landscapes at a continental scale because of the variation
at both landscape, patch scales and management scales, then it is optimal to use a standard
size, which then enables the direct comparisons to be made of relative heterogeneity. The 1
km square unit also enables internal spatial modelling of habitat patches and is sutiable for
scenario testing (Bunce et al. 1993).
In terms of the General Habitat Categories where a given category is in an optimum situation
for its development, it will occur extensively e.g. dwarf shrubby Life Forms below 0.05 m in
the Scandinavian mountains (Alpine North). In the Alps (Alpine South) it may occur as small
patches but not at all in most of Europe. Table 2 shows an example of how a given habitat
may occur in different landscapes. This complexity means that for any given survey, the
strategy has to be at a constant scale to enable comparison of relative extent to be made.
Subsequently, as described by Barr et al. (1993) further samples can be increased to reduce
standard errors, once the initial validation has been assessed. Alternatively, rare habitats, of
which the distribution is often known, can be targeted, either objectively using known
parameters e.g. coastline, or within patches identified in the field. In the latter case, statistical
estimates of extent cannot be made, but at least rare habitats can be monitored in this way.
Tests in BioHab of recording point information at intersections, as in the LUCAS project
programme (Gallego 2002) showed that not only spatial information was not produced but the
essential link to satellite imagery could not be made.
The Minimum Mappable Element (MME) is 400 m2 (with minimum dimensions of 5 x 80 m),
based on experience in the GB Countryside Survey (Barr et al. 1998) and in field trails in
1998 and 1999 in El Tiemblo near Madrid, Spain. This contrasts with the 250 000 m2 of the
CORINE
land
cover
map
and
2500
m2
of
the
Biopresss
Project
(http://www.creaf.uab.es/biopress/index2.htm). However, such detail is essential to express
the landscape ecological characteristics of many landscapes especially in the Mediterranean
region.
Within Bunce et al .(2005) detailed rules are provided for mapping e.g. for elements which
cross the boundaries of the 1 km2 and the use of interpreted aerial photographs to provide
initial boundaries for mapping. Some elements e.g. canals, motorways or major rivers may
be mapped as areal elements according to the rules, but may be subsequently allocated to
linear features by database management for specific objectives e.g. Haines-Young et al.
(2000). The fundamental principal is that disaggregated data are collected, so that subsequent
analysis can produce statistics that are sufficiently flexible to answer a range of policy
requirements.
Linear and Point Elements
In the majority of literature referred to above both linear and point features are largely
omitted. This also applies to most phytosociological studies e.g. in the Czech Republic over
30,000 relevees have been recorded, none of which are on linear or point features. However,
many studies have shown that especially in intensively managed agricultural landscapes
biodiversity has progressively become restricted to such situations e.g. Bunce & Hallam
(1993) and Hermy & de Blust (1997). Moreover, as intensification continues, so does the
pressure increase even in such limited areas e.g. Haines-Young et al. (2000) and Agger &
Brandt (1988).
Furthermore, many cultural landscapes are exceptionally rich in linear
features largely the product of management by managing the terraced landscapes of Crete and
the dense hedgerow network of the bocage. It is therefore essential not only to assess the
resources of linear and point elements in representative landscapes but also to monitor
change. As with areal elements a series of rules and protocols have therefore been developed
to maintain consistency and repatability as described by Bunce et al. (2005).
A linear element states that it must have a width of metre less than 5.0 m but more than 0.3 m
and must be longer than 30 m which is the Minimum Mappable Length (MML). Elements
that are smaller than 400m2 and shorter than 30 m can be recorded as points.
Linear habitats
often occur as complexes e.g. a fence, a ditch and a hedge, in which case instructions are
provided for mapping.
It is recognised that in many cultural landscapes the number of point features can be very
large e.g. trees along hedgerows and patches of rocks in fields. Two guidelines are provided
for recording such points:
1. Point features which add to landscape diversity, usually because they represent a
particular habitat which is generally absent from the surrounding area e.g. rock
outcrops or boulders in a grass field.
2. Point features which affect the ecological function of landscapes e.g. drinking places
in grasslands or weirs in watercourses which hinder migration.
However, a given survey may decide to omit point features and the procedure followed
should be recorded.
Validation
Previous experience in field workshops organised by the International Association for
Landscape Ecolgy (IALE) showed the difficulties in mapping theoretical classifications e.g.
EUNIS and CORINE Biotopes in the field, especially where continuous gradients were
concerned. It is therefore essential that rules were tested rigorously in the field in a variety of
situations.
Not only are many terms not defined but also sufficient detail is not provided for consistent
definition. In the case of Priority and Annex 1 Habitats especially, it was shown that further
information would be required e.g. on the natural distribution of critical species before they
could be recorded.
Excursions were made to diverse biogeographical locations e.g. in Almeria in south-east
Spain and inside the Arctic Circle in northern Norway, as shown in Figure 1.
These
excursions were selected to ensure that all major combinations of Life Forms were present
within the represented GHC’s. The rules were then discussed in the field and mapped in 1 km
squares in a range of workshops varying from Tessaloniki (Greece), Faro (southern Portugal),
El Tiemblo (central Spain), Poprad (Slovakia) and Uppsala (eastern Sweden). These practical
excursions and the exposure of the mapping procedures to external comments led to
progressive refinement of the field instructions and has ensured that the rules are sufficiently
robust to be applied across Europe. Some categories are rare and may never cover areas over
400 m2; hence the importance of recording linear features and points to express the variation
at the landscape level.
As the use of Life Forms is based on a regression model, the hypothesis can be tested,
although it is the substance of classical biogeography e.g. Walter (1973) and Woodward
(1991). The first such test was carried out in a valley in the Picos de Europea, north-west
Spain, which extended from broadleaved evergreen forest at 200 m to scree, rock and subalpine vegetation at 2500 m. Orthogonal regression, as described in the context of land
classification by Bunce et al. (1996), was used to calculate the correlation between mixture of
Life Forms in stratified random samples drawn from eight environmental strata using the
mean altitude of each strata as the independent variable. The correlation coefficient was 0.94
(6 df) and highly significant showing that at even a local scale the model was valid.
In the second test the data for elements used was that which had been collected on the
proportion of Life Forms from patches of > 400 m2 visited during the field excursions and
workshops.
These data can only act as a demonstrator as they are not collected from
representative random samples. The results are shown in Figure 2 and support the principle
of Life Forms and their relationship with the environment as expressed by Raunkiaer. The
results also show that in practice there are several significant dimensions e.g. from bare rock
to annual vegetation (high mountain to Mediterranean) and from grasses to spiny cusions
(temperate to Mediterranean). The axes are linked to the main environmental sectors of
Europe, as described by Metzger et al. (2005) and Jongman et al. (2005) showing that Alpine
North (Scandinavian mountains) and Mediterranean South (extreme southern Europe) were
isolated and the other zones clustered together showing the influence of management.
This analysis shows that Life Form combinations are more important than the individual
categories. They form complex relationships with the environment on the one hand but also
show modified patterns because of management by man on the other.
Stratification
The early part of this paper shows the level of decision making needed to make consistent
records as shown in Figure 1. Spatial and temporal aspects need to be concidered. The present
section deals with stratification to provide the former and the next section with the latter. The
collection of the field data must be based on a statistically sound sampling design so that it is
independent and unbiased, as outlined by Bunce et al 1996.
A stratification of framework has been constructed that optimises the selection of sampling
locations. Previous experiences (Bunce et al 1996) has used independent environmental
classifications derived from existing biogeoclimatic information. This approach has been
developed in Great Britain (Sheail and Bunce 2004) but also in Spain (Elena Rossello1997). It
is likely to be also efficient at a continental scale eg in Australia (Cawsey et al 2002).
The stratification (Metzger et al 2005, Jongman et al 2006) has been derived from a
statistical analysis of climatic and topografic data at a 1 Km square resolution. 13
environmental zones have been established linked hierachically to 84 environmental Strata.
This classification can be used to select the minimum of about 1400 1 km squares that would
be required to obtain statistically reliable estimates of the extent of GHC’s for Europe.
Existing data from objectively located samples can also be integrated where possible.
This method enables data from sample squares to be integrated first at the Stratum level and
then eventually for the whole of Europe by adding the figures from the Strata. Standard
statistical procedures, as defined by Barr et al 1993, can be used to obtain standard errors for
individual Strata for Europe. The data from the samples could also be linked to satellite
imagery based information e.g. the CORINE land cover map to improve the interpretation of
the categories.
Monitoring
Monitoring involves the temporal dimension referred to in Figure 1 and involves repeated
measurements at different time intervals. It is statistically essential to return to the same sites
to record changes (Barr et al. 1993, Brandt et al. 2003). This is the procedure followed in all
the major monitoring exercises in Europe. There are several networks already existing for
monitoring environmental changes employing various size units from 16 km squares down to
0.25 km squares, and sample extents that generally comprise between 0.2 and 0.3 % of the
spatial extent of the associated populations (Brandt et al. 2002b). Most of the field recording
is at the 1 km square level, as a compromise between detail and generality, and the BioHab
has therefore been based at this level.
The General Habitat Categories are specifically designed to be recorded consistently in order
to produce statistically robust estimates of extent. However, consistency becomes even more
imperative when the recording and mapping of change is concerned. The majority of field
habitat mapping exercises involve surveillance and are not designed to record change. More
stringent criteria are required in order to ensure that real change is recorded and not results
that are distorted by differences between observers and of recording technique. This requires
that emphasis in the re-survey be placed on registration of changes compared with the
recordings made previously e.g. the procedure used in Brandt & Agger (1988).
Thus,
information from the previous survey forms the basis for the field mapping and recording in
the re-survey, which is implemented as a check for change of each element recorded in the
previous survey. Change detection by independent surveys and subsequent data analysis is
time-consuming and can lead to uncertainties about whether the changes detected are real or
statistical artefacts.
Such monitoring has many advantages, especially when seen in the long-term, as it allows
checking of the quality of each of the surveys. Each registration of a change generates the
question: is it a real change? Or is re-evaluation of the earlier registration required? This
permits a higher degree of confidence in the data as the number of surveillance events
increases. The result of this procedure is that the monitoring has not only become more
reliable, due to better registration techniques, but also the editing of former registrations has
added to the quality. In fact, a considerable part of the time needed for the refinement of the
database in a Danish national landscape monitoring system made every 5 years from 1981 to
2001 has been devoted to the systematic control of all detected changes back in time (Brandt
et al. 2002a). Such a rigorous control is necessary, since landscape monitoring relies on the
detection of small changes and using this procedure guarantees that the changes have actually
taken place. The statistical confidence that can be attached to the measures can however be
low if changes are rare.
There is much experience in applying such methodology in the detection of change e.g.
Northern Ireland, Denmark and Great Britain, and in interpreting changes from aerial
photographs e.g. Spain, Sweden and The Netherlands. One of the key elements of this
approach is the detection and evaluation of alternations within habitat types, e.g. new forestry
planted on blanket bogs is negative, but is positive if taking place on arable land (Figure 3).
Data Quality and Quality Control
Whilst it is recognised that the information on management is more difficult to record than the
GHC’s, Kirby et al. (2005) have shown that such data can reveal significant changes if the
sample is sufficiently large. Also it is not necessary to always have data collected in the same
sites, as Bunce et al. (1999) showed for moths and freshwater invertebrates. In addition data
from socio-economic surveys could be linked via the Environmental Strata of Metzger et al.
(2005) by providing the boundaries of the strata to third parties who hold confidential
information.
The various sets of qualifiers described above are considered provisional, and additional
categories could be added by consultation with regional experts. It is also recognised that the
descriptions would need to be improved to make them fully understood across Europe.
Further details could also be added on landscape character and geomorphological features.
Barr et al. (1993) and Kirby et al. (2005) both showed the importance of time of year that the
records are made. Detailed contingency plans are therefore needed to ensure that the data of
survey in individual regions are not only time for phenology but also for variations between
years. This is most likely to be important in the Mediterranean region where variation is
generally greater between years than elsewhere.
Quality control and assurance are essential to ensure reliable data. Quality control procedures
involve checking field surveyors identification and mapping skills by experienced staff
actually in the field. Whilst this is most important during the early stages of the project, it is
also essential throughout and should include independent random checks of individual
samples. This was carried out in the UK Countryside Surveys in 1993 and the results
compared using primary codes of major habitats, comparable to GHC’s and land use (Barr et
al. 1993). Analysing these codes showed a correspondence of 84% (95% in fields; 71% in
open vegetation). The only directional bias was the distinction between heaths and bogs,
which are now included only as qualifiers. The differences were mainly due to the actual data
of the survey, which has led to future surveys being carried out at the same date as the
baseline. The secondary codes (comparable to qualifiers in the present paper) had 75%
correspondence. The conclusion was that the data were sufficiently robust to detect change
and implicitly, that the approach described in the present paper will also be able to detect
change.
Discussion
The Policy Background
Since the Rio Convention on Biological Diversity was signed by 150 governments after its
agreement in 1992, there have been a series of policy initiatives relating to biodiversity. One
of the issues in conserving in situ biodiversity is the conservation of threatened habitats as
laid down in the European Habitats and Species Directive (Council Directive 92/34/EEC). A
full review of all initiatives and actions is beyond the scope of the present paper, which
primarily concerns habitats within the biodiversity hierarchy but could be extended to
vegetation, species and genetic levels of biodiversity.
Within the Directive, the conservation of habitats is elaborated in article 2, 3 and 4, with an
extensive list of habitats of European importance and priority habitats for conservation given
in Annex 1. This list of habitats was derived form the CORINE Biotope project which was
initiated in 1986 but reported in 1991 (Devillers et al. 1991) and led to the list of habitats in
the Directive, within which certain Priority Habitats were identified for legal protection. This
list has been progressively modified by expert consultation following addition of new
accession states e.g. Finland and Austria. The Directive has been subsequently augmented by
other initiatives e.g. the Gothenbug Commitment by the EU, that biodiversity decline should
be halted by 2010. All of these iniatives are needed in support of a baseline database, using
extant information and both for assessing the extent of habitats and as well as new data and
subsequent monitoring.
Such a baseline is currently lacking and projects such as MIRABEL (Petit et al 2001) have
only been able to use expert judgement for assessing the distribution and extend of European
habitats. Mucher et al 2004 have used the descriptions in Annex 1 to derive rules using
existing databases to predict distribution of habitats. However, many of the descriptions do
not contain enough detail for mapping.
Over recent years the Natura 2000 series of sites has been set up as the major initative to
maintain habitats and their associated biodiversity. The selection of these sites is primarily
based on Annex 1 Habitats but also reflects national priorities. However, inevitably such sites
can only cover a small proportion of the European land surface (probably about 10%); outside
their boundaries there is limited or no protection of habitats.
Nevertheless, the non-
designated land not only contains a high proportion of the total wildlife resource, but is
actually what most people experience in everyday life. It is therefore essential for a panEuropean procedure to adequately define rules to record habitats on such land, as well as the
mainly semi-natural habitats of Natura 2000 sites, hence the development of the present
procedure. It is also essential to provide an objective baseline as described in the present
paper against which the effectiveness of protection within Natura 2000 sites can be measured.
None of the existing procedures adequately covers the complexity of landscapes nor the
recognition of the spatial heterogeneity across Europe.
Although the Natura 2000 network is based on Annex 1 and Priority Habitats, some countries
e.g. Spain have interpreted the remit in a broad way. In addition the interpretation of the
habitat types also varies between countries. This is partly because the types themselves are
derived from a mixture of general information and phytosociological classes and also because
they are in response to different national priorities.
In general, the individual species provided in the Annex 1 descriptions should not be used
alone to identify habitats, but rather a complex of associated species, although again this is
difficult to apply consistently. The current database in Natura 2000 is difficult to use, because
inconsistencies may be due to national interpretations or even data errors. There are also gaps
in the series e.g. Calthion in The Netherlands, which have been added to the nearest class but
this has not been carried out consistently across the EU.
Many of the types are of such restricted distribution that they cannot be picked up by random
or grid samples. Instead a series of individual stratifications and masks eg. Coastal would be
required. An initial trial has shown how the GHC’s could provide the necessary framework
by reducing the uncertainty in the identification of Annex 1 habitats, for which at present
there is no key and inadequate descriptions eg 9260 Castanea sativa woods have three lines
whereas 6230 species rich Nardus grasslands have three pages . A possible approach is to
develop an expert system which can then be circulated to regional experts to add their local
knowledge.
References
Agger, P. & Brandt, J. 1988. Dynamics of small biotopes in Danish agricultural landscapes.
Landscape Ecology 1: 227-240.
Barr, C.J., Bunce, R.G.H., Clarke, R.T., Fuller, R.M., Furse, M.T., Gillespie, M.K., Groom,
G.B., Hallam, C.J., Hornung, M., Howard, D.C. & Ness, M.J. 1993. Countryside Survey
1990: main report volume 2. Department of the Environment, London, 174pp.
Barr, C.J.
1998.
Countryside Survey 2000 Field Handbook.
Centre for Ecology and
Hydrology, Grange-over-Sands, U.K.
Braun-Blanquet, J. 1928. Pflanzensoziologie: Grundzuge de Vegetationskunde. Springer
Verlag, Berlin, 330pp.
Bunce, R.G.H. & Shaw, M.W. 1973. A standardised procedure for ecological survey.
Journal of Environmental Managmeent1: 239-258.
Bunce, R.G.H. & Hallam. C.J. 1991. Diversity of vegetation in British landscapes. In:
Pineda, F.D., Casado, M.A., de Miguel, J.M. & Montalvo, J. (eds), Biological Diversity.
Fundacion Ramon Aceres, Madrid, pp 77-81.
Brandt, J.J.E., Bunce, R. G. H., Howard, D. C. & Petit, S. 2002a. General principles of
monitoring land cover change based on two case studies in Britain and Denmark. Landscape
and Urban Planning 62: 37-51
Brandt, J., De Blust, G., & Wascher, D. 2003. Monitoring multifunctional terrestrial
landscapes. In: Brandt, J. & Vejre, H. (eds): Multifunctional Landscapes. Vol. II: Monitoring,
Diversity and Management. WITPress, Southhampton, pp. 75-84.
Brandt, J., Holmes, E. & Ravn, H. P. 2002b. Biodiversity and trees outside forests: The case
of Denmark. In: Puumalainen, J., Kennedy, P., & Folving, S. (eds). Forest Biodiversity –
Assessment Approaches for Europe. EUR Report 20423 EN. Joint Research Centre (DG
JRC). Ispra. European Communities. Pp. 80-85.
Bunce, R.G.H., Howard, D.C., Hallam, C.J., Barr, C.J. & Benefield, C.B. 1993. Ecological
consequences of land use change. Countryside Survey 1990, Volume 1. Department of the
Environment, London, 129pp.
Bunce, R.G.H., Barr, C.J., Clarke, R.T., Howard, D.C. & Lane, A.M.J. (1996a). Land
classification for strategic ecological survey. Journal of Environmental Managmeent 47: 3760.
Bunce, R.G.H., Smart, S.M., van de Poll, H.M., Watkins, J.W. & Scott, W.A.
1999.
Measuring Change in British Vegetation. ECOFACT Volume 2. NERC-CEH, pp 144.
Bunce, R.G.H., Carey, P.D., Elena-Rossello, R., Orr, J., Watkins, J.W. & Fuller, R. 2002. A
comparison of different biogeographical classifications of Europe, Great Britain and Spain.
Journal of Environmental Management 65: 121-134.
Bunce, R.G.H., Groom, G.B. Jongman, R.H.G., Padoa-Schippa, E. (eds). 2005. Handbook
for Surveillance and Monitoring of European Habitats. First Edition. Alterra, Wageningen,
107pp.
di Castri, F., Goodall, D.W. & Specht, R.L. (eds). 1981. Mediterranean-type ecosystems of
the world. Elsevier, Amsterdam, 52pp.
Cawsey, E.M., Austin, M.P. & Baker, B.L. 2002. Regional vegetation mapping in Australia: a
case study in the practical use of statistical modelling. Biodiversity and Conservation, 11,
2239-2274.
Clapham, A.R., Tutin, T.G. & Warburg, E.F. 1952. Flora of the British Isles. Cambridge
University Press, 1591pp.
Davies, C.E. & Moss, D. 2002. EUNIS Habitat Classification. Final Report to the European
Topic Centre of Nature Protection and Biodiversity, European Environment Agnec y, 125pp.
Den Boer, P.J. & Van Dijk, T.S.
1994.
Carabid beetles in a changing environment.
Wageningen Agricultural University Papers, 30pp.
Devillers, P., Devillers-Terschuren, J. & Ledant, J.P. 1991. CORINE biotopes ma nual.
Habitats of the European Community. Data Specifications – Part 2. Office for Official
Publications of the Euroepan Community, Luxembourg, 300pp.
Devillers, P. & Devillers-Terschuren, J. 1996. A classification of paearctic habitats and
preliminary habitats in Council of Europe Member States. Report to the Council of Europe
Convention on the Conservation of European Wildlife and Natural Habitats, 268pp.
Elena-Rossello, R. (ed).
1997.
Classifcation Biogeoclimatica de Espana Peninsular y
Baleaur. M.A.P.A., Madrid, 448 pp.
Ellenberg, H., Weber, H.E., Dull, R., Wirth, V., Werner, W., Pauliben D.
1992.
“Zeigerwerte von Pflanzen in Mitteleuropea”. Scripta Geobotanica 18 (2nd edition).
European Communities. 1992. Council Directive 92/43/EEC on the conservation of natural
habitats and of wild fauna and flora. Text and Annexes of the Habitat Directive. Office for
Official Publications of the European Communities, 21 May 1992.
Fjellstad, W.J., Dramstad, W.E., Strand, G.H. & Fry, G.L. A. 2001. Heterogeneity as a
measure of spatial pattern for monitoring agricultural landscapes. Norwegian Journal of
Geography 55: 71-76.
Fernandez, M. (coord).
2003.
Man ual de interpretaction de los habitats de Espana.
TRAGSA, Ministerio de Medio Ambiente, Madrid, Vols 1 & 2, 479pp.
Fuller, R.M. & Parsell, R.J. 1990. Classification of TM imagery in the study of land use in
lowland Britain: practical considerations for operational use. International Jorunal of Remote
Sensing 11: 1901-1917.
Gallego, J. 2002. Building Agro Environment Indicators. Focussing on the European area
frame survey LUCAS. EUR Report 20521N. European Commission, Ispra.
Grabherr, G., Gottfried, M. & Pauli, H. 1994. Climate effects on mountain plants. Nature
369: 448-448.
Green, R.H. 1979. Sampling design and statistical methods for environmental biologists.
John Wiley & Sons, New York, 257pp.
Greig-Smith, P. 1964. Quantitative plant ecology. 2nd edition. Butterworth, London, 256pp.
Haines-Young, R.H., Barr, C.J., Black, H.I.J., Briggs, D.J., Bunce, R.G.H., Clarke, R.T.,
Cooper, A., Dawson, F.H., Firbank, L.G., Fuller, R.M., Fuirse, M.T., Gillespie, M.K., Hill,
R., Hornung, M., Howard, D.C., McCann, T., Morecroft, M.D., Petit, S., Sier, A.R.J., Smart,
S.M., Smith, G.M., Stott, A.P., Stuart, R.C. & Watkins, J.W. 2000. Accounting for nature:
assessing habitats in the UK countryside. DETER, London, 134pp.
Hermy, M. & de Blust, G. 1997. Punten en Lijnen in het Landschap. Schuyt en Co,
Haarlem, 335pp.
Hill, M.O., Roy, D.B., Mountford, J.O. & Bunce, R.G.H. 2000. Extending Ellenberg’s
indicator values to a new area: an algorithmic approach. Journal of Applied Ecology 37: 315.
Intergovernmental Group on earth Observation. 2004. The Global Earth Observation System
of Systems (GEOSS). 10-Year Implementation Plan.
JNCC. 1990. Handbook for Phase 1 Habitat Survey: a technique for environmental audit. 2
volumes. Joint Nature Conservancy Council, Peterborough, Vols 1 & 2, 143pp.
Jongman, R.H.G., Bunce, R.G.H., Metzger, M.J., Mucher, C.A. & Howard, D.C. 2006. A
statistical Environmental Classification of Europe: objectives and applications. Landscape
Ecology 21: 409-419.
Kirby, K.J., Smart, S.M., Black, H.I.J., Bunce, R.G.H., Corney, P.M. & Smithers, R.J. 2005.
Long term ecological change in British woodland (1971-2001). A resurvey and analysis of
change based on the 103 sites in the Nature Conservancy ‘Bunce 1971’ woodland survey.
English Nature Resarch Report No 653, English Nature, Peterborough, 139pp.
Lambert, J.M. 1972. Theoretical models for large-scale vegetation survey. In: Jeffers, J.N.R.
(ed), Mathematical models in ecology. Blackwell, Oxford, pp. 87-109.
Metzger, M.J., Bunce, R.G.H., Jongman, R.H.G., Mucher, C.A. & Watkins, J. 2005. A
climatic stratificiation of the environment of Europe. Global Ecology and Biogeogrtaphy 14:
549-563.
Mucher, C.A. Hennekens, S.M., Bunce, R.G.H. & Schaminee, J.H.J. 2004, Mapping
European habitats to support the design and implementation of a PAN-European Network,
Alterra 952, Alterra, Wageningen.
Oberdorfer, E., Mueller, T., Korneck. D., Lippert, W., Makgraf-Dannenberg, I.
1990.
Pflanzensoziologische Exkursionsflora. 6th edition. Stuttgart Ulmer, 1050pp.
Petit, S., Firbank, L. Wyatt, B. & Howard, D., 2001, MIRABEL, Models for integrated
assessment of biodiversity in European landscapes, AMBIO 30, 81-88.
Pignatti, S. 1982. La flora d’Italia. Edagricole, Bologna, 2324pp.
Pyatt, D.G. 1998, Correlation of European Forest Site Classifications. EU Concerted Action
Final Report, Forest Reseach Northern Station, Roslin.
Quezel, P. & Barbero, M. 1982. Definition and characterization of Mediterranean-type
ecosystems. Ecologia Mediterranea 7: 15-29.
Raunkiaer, C.
1907.
Planterigets livsformer og deres Betydning for Geografien.
Munksgaard, Copenhagen, 102pp.
Raunkiaer, C. 1934. The Life Forms of plants and staitsical plant geography, being the
collected papers of C. Raunkiaer. Clarendon, Oxford, 632pp.
Rodwell, J.S., Schaminee, J.H.J., Mucina, L., Pignatti. S., Dring, J. & Moss, D. 2002. The
Diversity of European Vegetation. An overview of phytosociological alliances and their
relationships to EUNIS habitats. Ministry of Agriculture Nature Management and Fisheries,
The Netherlands and European Environment Agency, Wageningen, 168pp.
Sheail, J. & Bunce, R.G.H.
2003.
The development and scientific principles of an
environmental classification for strategic ecological survey in Great Britain. Environmental
Conservation 30¨147-159.
Tuxen, R.
1937.
Die Pflanzengesellschaften Nordwestdeutschlands.
Mitteilungen der
Florstisch-soziologischen Arbeitsgemeinschaft in Niedersachsen 3: 1-170.
Walter, H.
1973. Vegetation of the earth in relation to climate and eco-physiological
conditions. Springer Verlag, New York, 237pp.
Woodward, F.I. & Rochefort, L.
1991
Sensitivity analysis of vegetation diversity to
environmental change. Global Ecology and Biogeography Letters 1: 7-23.
Table 1. Overview of super categories of Habitat types and the related General Habitats
categories (Bunce et al. 2005).
Table 2. Examples of species with varying degrees of plasticity.
Dwarf
Shrubby
Low
Mid
Tall
Forest
Chamaephytes
Chamaephytes
Phanerophytes
Phanerophytes
Phanerophytes
Phanerophytes
DCH
SCH
LPH
MPH
TPH
FPH
0.01 – 0.05
0.05 – 0.30
0.30 – 0.60
0.60 – 2.00
2.00 – 5.00
> 5.00
Winter deciduous
Salix herbacea
X
Salix serphyllifolia
X
Betula nana
X
X
Vaccinium myrtillus
X
X
Myrica gale
X
X
Rosa pimpinellifolia
X
X
Alnus viridis
X
X
X
Amelanchier ovalis
X
X
X
Salix cinerea
X
X
X
X
Fragula alnus
X
X
X
X
Quercus petraea
X
X
X
X
X
Crategus monogyna
X
X
X
X
X
Evergreen
Dryas octopetala
X
Vaccinium oxycoccus
X
Helianthemum
X
alpestre
Arctostaphylos
uva-
X
X
X
X
ursi
Vaccinium vitis-idea
Thymus vulgaris
X
Lavandula stoechas
X
Siderits syriaca
X
Helichrsym stoechas
X
Daphne laureola
X
Rubus idaeus
Vaccinium
X
X
X
X
X
X
X
X
uliginosum
Empetrum nigrum
X
Table 3. Main rules for determining General Habitat Categories and mapping units.
GHC’s
Mapping
Units
A GHC has to be determined by one field visit, or from extant data at a scale of at least
1:10,000, which must be made in an appropriate time window for a given region, i.e.
either side of the period of maximum biomass.
GHC’s must be mutually exclusive and together cover the complete land surface of
Europe, including water bodies.
GHC’s must be a common denominator for comparison between countries using extant
data and classes in current use wherever possible.
GHC’s must be distinctive and recognisable.
There must be explicit rules to define GHC’s.
Differences in management are recorded as qualifiers and are not in the definitions of
GHC’s but might influence the delineation of mapping units
Habitats are not defined on the basis of biogeographic regions because of difficulties of
maintaining consistency due to the lack of adequate definitions of the multiplicity of
terms. Any biogeographical term that can be determined consistently can be attached to
GHC’s through database management.
Individual species are not used to identify GHC’s because of vicarious species and
differences in species behaviour in contrasting biogeographical regions. However the
use of indicator species to identify environmental qualifiers is useful.
The recommended basic survey area is 1 km square within which areal, linear and point
elements are recorded. In complex landscapes 0.25 km squares may be appropriate. The
key to the General Habitat Categories can however be applied to any extant data or for
general recording in the field.
The Mimimum Mappable Element (MME) for an areal element is 400 m square with
minimum dimensions of 5 x 80 m; if it is smaller than 5 m ( measured as seen in a
vertical perspective) the element is recorded as a linear element with a Minimum
Mappable Length (MML) of 30 m. Elements that do not pass the minimum size criteria
for either areal or linear elements can be mapped and recorded as point elements or as
proportions of a larger element.
Areal elements with a total extent that passes the MME criteria and lie across the edge of
the survey square should be recorded as areal elements even if the part of the element
that is within the survey square is below 400 m square. Linear elements of more than 30
m crossing the edge should be recorded if more than 20 m are situated within the area.
Roads canals, and broad rivers may be linear elements, but if they are over 400 m square
within the survey area and at leeat 5 m wide,they are mapped as aerial elements.
(Subsequent database analysis can analyse these as linear elements, if required).
Table 4. Procedures and rules for field recording as developed in the BioHab project (Bunce
et al. 2005).
1
Surveyors are provided with lists of GHC’s and qualifiers to be used to describe
each mapped element (area, line or point) in the survey area.
2
Non-standard secondary codes can be used for site and management qualifiers if the
observed site or qualification is not covered by the standard site and management
qualifier code listed.
3
The surveyor should record data of areal elements on one recording sheet and data
of linear and point elements together on another recording sheet. A third sheet is
provided for background information on the survey square.
4
Elements are assigned alpha codes that are the same on the map and on the
corresponding recording sheet. Elements with an identical combination of sheet
codes can be assigned the same alpha code.
5
The total cover is estimated as from a vertical perspective and the mapping of areal
elements adds to 100% of the land surface. The entire survey area must be mapped,
even the small corners of the square.
6
Multiple vegetation layers e.g. within forests are not recorded, but could be subject
of an additional module within regional surveying activities.
7
Point elements are recorded if they are considered significant in the landscape
context. It must be made explicit how these have been recorded, so that they can be
monitored effectively.
Table 5. Field recording format for European wide recording including qualifiers.
Code
Field 1
Field 2
Field 3
Field 4
Field 5
Field 6
Field 7
Field 8
α
General Habitat
Global/
Site
Man.
Life Form/Species
Pan
Region
Phyto-
Category
Env.
Qualifier
Qualifier
Europ.
al
sociolo
Class
Class
gy
-1
-1
-1
Qualifier
A
TPH/DEC/CON
NEW
5.3
0
Life Form
%
Species
%
421
TPH/DEC
20
Sor auc
80
429
TPH/CON
10
Pica bi
100
MPH/EVR
60
Rub fru
100
FPH/CON
10
Pica bi
100
Table 6. Definitions of super categories of habitat
URBAN/CONSTRUCTED The definition of urban and constructed land codes covers “elements
associated with built structure and routes of communication. Linear
elements which are immediately adjacent to an urban element are
not be recorded, except for roads”. Land is defined as urban, when
it “is an area of ground that is associated with a building and which
has a use linked to that building e.g. garden”. In most European
countries there are clearly marked boundaries around urban land and
creation areas e.g. The Netherlands, Spain and Belgium, whereas in
other countries e.g. Austria, Estonia and Norway there may not be
actual physical boundaries around the houses.
Then the urban
boundary should be drawn around the grounds of a building where
the management intensity changes from that of a gardening
character to more extensive management types.
CULTIVATED
Crops are mainly the product of plant breeding, but also of native
specie such as walnut. Wild species collected from semi-natural
vegetation are excluded.
SPARSELY
Elements which have less than 30% cover of vegetation, excluding
VEGETATED
saxicolous, lichens and bryophytes.
HERBACEOUS
Herb vegetation if the cover of shrubs and trees is below 30%. They
can cover all kinds of vegetation from water plants (helophytes) to
bogs (Cryptogams) and grasslands (leafy hemicryptophytes and
caespitose hemicryptophytes).
SHRUBS AND TREES
Most of these habitat categories are woody – the term usually used
in habitat classifications – but some chamaeophytes e.g. Phagnalon
spp., Artemisia spp., Asparagus spp. Do not have secondary
ligneous woody thickening in strict botanical terminology.
However, these genera have a shrubby form and have penennating
buds above ground level. The woody trees and shrubs refer to
individual plants and Life Forms. In the landscape groups of trees
and shrubs combine to form forest and shrub habitats.
Table 7. Definitions of moisture regimes.
Waterlogged/water-
Water table at the surface with standing water from between 50 and
saturated
70% of the year or with the soil completely saturated. Only small
patches may become wet in mid summer.
Wet
Water table < 40 cm under the surface and soil containing free
water for most of the year.
Seasonally wet
Water table variable at the surface and waterlogged for the winter
months or spring flooding season, becoming wet or mesic during
the summer period.
Mesic
Water table 40-100 cm of the surface, available water during most
of the non summer period, may dry out during the mid summer
period.
Dry
Water table > 100 cm of the surface, water available only during
some periods.
Very Dry
Water table > 100 cm of the surface, dry throughout most of the
year with only short mesic periods.
Xeric
Water table > 100 cm of the surface, dry throughout the year except
in isolated rain events.
Table 8. Matrix of nutriant levels (Ellenberg values).
Aquatic
Water-
Seasonally
logged
wet
Wet
Mesic
Dry
Very
Xeric
Dry
Eutrophic
1.1
2.1
3.1
4.1
5.1
6.1
7.1
8.1
Acid
1.2
2.2
3.2
4.2
5.2
6.2
7.2
8.2
Neutral
1.3
2.3
3.3
4.3
5.3
6.3
7.3
8.3
Basic
1.4
2.4
3.4
4.4
5.4
6.4
7.4
8.4
Saline
1.5
2.5
3.5
4.5
5.5
6.5
7.5
8.5
Figure 1. Distribution of the main visits and workshops in the field validation.
Figure 2.
Biplot of General Habitat Categories (GHC’s) coded as in Section 11 and
Environmental Zones (EnZ's) resulting from Canoncial Correspondence Analysis (DCA).
The first two axes are shown. Eigenvalue of Axis 1 is 0.42; eigenvalue of Axis 2 is 0.35. The
Zones are Alpine North; ALN, Alpine South: ALS, Atlantic North: ATN, Altantic Central:
ATC, Lusitanian: LUS, Boreal: BOR, Nemoral, NEM, Continental: CON, Pannonian: PAN,
Mediterranean North: MDN, Mediterranean Mountains: MDM, Mediterreanean South: MDS.
Figure 3. Diagram of principal potential flows between Life Form categories.